Evaluation of oxidative stress after fractures.

A preliminary study
Ganesh PRASAD, Mandeep S. DHILLON, Madhu KHULLAR, Onkar N. NAGI

In a preliminary attempt to see if oxidative stress

occurs after major fractures, we evaluated two
groups of patients sustaining a single fracture
(Group A) and multiple fractures (Group B), and
compared them with healthy controls (Group C).
Indirect evaluation using plasma was done, as serial
samples directly from bone could not be taken in
humans. We measured plasma levels of malonyldialdehyde (MDA), which depicts the lipid peroxide
content, and the unstable nitric Acid (indirectly
through measuring reactive nitrogen intermediates
and citrulline), serially over a four-week period.
The endogenous ferric reducing anti-oxidant power
assay (FRAP) was also done. All these have been
proven to be representative of oxidative stress in
other situations. We noted significant changes in
these values, peaking by the 2nd and 3rd weeks post
fracture, and coming down by the 4th week. This
implies that oxidative stress does occur after a fracture ; the initial few days are eventless, probably
because the fracture causes a local region of
ischaemia. Oxidant levels rise by the 2nd and 3rd week,
perhaps due to callus formation and angiogenesis,
which results in reperfusion at the fracture site.
Oxidative stress may also be proportional to the
number of bones fractured, as Group B, with multiple fractures had more elevated values. The antioxidant levels also behave similarly to combat the detrimental effect. The pro and antioxidants values then
gradually decline by the 4th week, probably because
by then the bone starts to organize. A better understanding of these mechanisms may help in defining
the role of oxidative stresses after fracture and perhaps better define the role of antioxidants in helping
fracture healing.

Acta Orthopdica Belgica, Vol. 69 - 6 - 2003

INTRODUCTION
In the last twenty years, more and more emphasis has been laid on the oxygen free radicals as a
major common final pathway of tissue injury in
different organ systems. Oxidative stress is referred
to as the chain of oxidative events that leads to
increased production of reactive oxygen species
which cause tissue injury. Following a fracture,
oxidative stress injury may be caused by an
ischemia-reperfusion mechanism (14, 17, 20). The
first three days of fracture healing may be compared to the ischemia period, where no oxidative
stress injury occurs. After this, in the stage of callus
formation, in addition to fibroblast and collagen
cells, new capillary vessels with other inflammatory cells increase the production of oxygen free

From the Postgraduate Institute of Medical Education and

Research, Chandigarh, India.
Ganesh Prasad, MBBS, MS : Senior Orthopaedic Resident.
Mandeep S. Dhillon, MS, MNAMS : Additional Professor.
Department of Orthopaedics.
Madhu Khullar, MSc, PhD : Additional Professor.
Department of Experimental Medicine and Biotechnology.
Onkar N. Nagi, MS, MSc (Oxon), FAMS : Professor and
Chairman.
Department of Orthopaedics.
Postgraduate Institute of Medical Education and Research,
Chandigarh, India.
Correspondence : M S Dhillon, 1090/2 Sector 39-B,
Chandigarh, 160036 India. E mail : msdhillon@sancharnet.in ;
mandyrima@hotmail.com.
2003, Acta Orthopdica Belgica.

547

EVALUATION OF OXIDATIVE STRESS

radicals. These radicals may cause oxidative injury

to the fractured bone as has been seen in other tissues with reperfusion injury (8). As the understanding of oxygen free radical damage in different
organ systems is evolving, it is perhaps justifiable
to attempt to define the nature of oxidative stress in
bone trauma and to evaluate if there are any deleterious effects on fracture healing (8, 12, 24).
Due to the attack of oxygen free radicals on the
lipid component of membrane, the lipid peroxide
content is elevated. To evaluate this, estimation of
levels of many intermediate lipid peroxides and
their end-products has been used to indirectly evaluate the oxidative stress. The most reliable indicators are Malonyldialdehyde (MDA) or Thiobarbituric acid reducing substances (TBARS) (8, 12, 24).
The involvement of Nitrous oxide (NO) as an effector molecule in many physiological and pathological situations has been firmly established. Most of
the previous work relating to oxidative stress after
a fracture has been based in the experimental setting (12, 24) with minimum data in humans (2). Since
direct bone samples cannot be taken serially in
humans, we attempted to study the oxidative stress
after fractures by evaluating the blood levels of
nitric oxide (NO) and Malonyldialdehyde (MDA).
NO is chemically unstable, and hence its levels
were monitored indirectly by measuring the levels
of reactive nitrogen intermediates (RNI) and citrulline (cit) ; we also evaluated the endogenous antioxidant status by ferric reducing anti-oxidant
power (FRAP) assay at weekly intervals.
PATIENTS AND METHODS
Sixty patients between the ages 18-60 years were
taken up for this study. All cases had purely musculoskeletal injuries. The cases were divided into two
groups of 30 patients each, with Group A having patients
with an isolated femur fracture, and Group B consisting
of patients with a femur fracture associated with another long bone fracture. Twenty human volunteers (Group
C) with no ailment served as controls, and had the tests
done in order to form a standard of comparison.
Five milliliters of heparinised blood was drawn from
each patient and evaluated for reactive nitrogen intermediates(RNI), citrulline (cit), malonyldialdehyde (MDA)
and total endogenous anti-oxidant activity. Blood sam-

ples were taken on days 1, 7, 14, 21 and 28 post

fractures.
Estimation of NO as RNI was based on the principle
that nitrite in the sample reacts with Griess reagent to
form a purple ozo dye which was measured spectrophotometrically by Greens method (8, 9). The estimation of
Citrulline was determined calorimetrically using
diacetyl monoxime by the Boyde and Rahmatullah
method (5), estimation of MDA was done by the Stocks
and Dormandy method (22) and the measurement of total
anti-oxidant activity was carried out by the FRAP
assay (4, 7).
One way analysis of variance (ANOVA) was used to
determine the statistical significance among groups and
the paired t test to detect a significant difference
between study groups. Statistical significance was set at
a level of p < 0.05

RESULTS
In the 60 patients evaluated after trauma, sex distribution was similar (Group A -27M/3F and Group
B -28M/2F), reflecting the disparity of involvement
in road accidents between men and women. The
control group comprised of 10 males and
10 females. The mean ages in groups A, B and
controls (C) were 30.07, 33.77 and 32.85 years
respectively. There was no statistically significant
correlation between the oxidative stress and either
sex or age.
There was a significant rise in the RNI levels
during the 7th day and a peak during the 14th day in
both groups A and B patients as compared to the
controls (table I). The increase in group B was
significantly more as compared to group A. In
group A these values decreased almost to the initial
values as on day 1 by the 28th day. In group B,
the values decreased to a level less than those as on
day 1.
There was a significant rise in the Citrulline
levels during the 7th day and a peak during the
14th day in both groups A and B patients as compared to the controls (table II). The increase in
group B was again significantly more as compared
to group A. These values decreased almost to the
initial values by day 21. On the 28th day, these values also decreased to a level less than those noted
on day 1.

Acta Orthopdica Belgica, Vol. 69 - 6 - 2003

548

G. PRASAD, M. S. DHILLON, M. KHULLAR, O. N. NAGI

Table I RNI values in the control (C) and study groups (A & B) on days 1, 7, 14, 21, 28 after injury

Control (nmol/ml) (n = 20)

A (nmol/ml) (n = 20)
B (nmol/ml) (n = 20)

Day 1
(mean SE)

Day 7
(mean SE)

Day 14
(mean SE)

Day 21
(mean SE)

Day 28
(mean SE)

358.70 30.50
230.39 21.10
313.51 41.08

282.19 27.34*
434.76 43.33*

540.11 38.89**
791.95 81.27**

344.83 28.32
367.63 31.88

226.15 22.80
237.16 18.60

p < 0.05, significant **p < 0.01, very significant.

Table II. CIT values in the control (C) and study groups (A & B) on days 1, 7, 14, 21, 28 after injury

Control (nmol/ml) (n = 20)

A (nmol/ml) (n = 30)
B (nmol/ml) (n = 30)

Day 1
(mean SE)

Day 7
(mean SE)

Day 14
(mean SE)

Day 21
(mean SE)

1251.94 68.35
1261.91 113.38
1259.15 40.08

1583.43 136.14* 1889.34 111.24** 1182.39 65.16

1396.25 56.20* 2088.97 104.72** 1295.13 56.13

Day 28
(mean SE)
853.61 57.65
983.60 34.06

p < 0.05, significant, **p < 0.01, very significant.

Table III. MDA values in the control (C) and study groups (A & B) on days 1, 7, 14, 21, 28 after injury

Control (mol/ml) (n = 20)

A (mol/ml) (n = 30)
B (mol/ml) (n = 30)

Day 1
(mean SE)

Day 7
(mean SE)

Day 14
(mean SE)

Day 21
(mean SE)

Day 28
(mean SE)

0.0063 0.0013
0.47 0.14
0.60 0.12

0.42 0.13
0.62 0.14

1.17 0.20**
1.90 0.34**

0.34 0.11
0.66 0.17

0.08 0.04
0.56 0.12 **

p < 0.01, very significant.

There was a significant rise in the MDA levels

by the 14th day in group A patients (table III). In
group B, there was slight increase in values on day 7
and a significant rise in values on day 14. The
increase in group B was again significantly more as
compared to group A. By the 21st and 28th days, MDA
levels decreased to less than values seen on day 1.
There was a significant rise in the FRAP levels
by the 14th day in both groups A and B patients as
compared to the controls (table IV), with a significantly more increase in group B levels. By the 28th
day, FRAP values again decreased to a level less
than those as on day 1.
DISCUSSION
Growing evidence strongly suggests that antioxidants have the power to prevent diseases such as
Acta Orthopdica Belgica, Vol. 69 - 6 - 2003

cancer (19), coronary artery disease (1) and

cataract (6). Antioxidant vitamins may improve
immune system function and may even delay some
of the effects of aging (3, 16). This is probably due
to their ability to intercept and extinguish free radicals. They also have some protective role in diabetes (6, 16), Alzheimers disease (3, 21) and
Parkinsonism. Radiation, pollution, smoking, alcohol addiction and obesity have adverse effects on
the biological systems and the literature (6, 16) has
supported some benefit of antioxidants to counter
these adverse effects. So much is talked about
antioxidants and their role in physiological and
pathological conditions that it made us think of the
importance of these in our subset of patients who
sustained fractures. This study was designed in an
effort to evaluate oxidative stress that occurs after
sustaining a fracture.

549

EVALUATION OF OXIDATIVE STRESS

Table IV. FRAP values in the control (C) and study groups (A & B) on days 1, 7, 14, 21, 28 after injury

Control (mol/L) (n = 20)

A (mol/L) (n = 30)
B (mol/L) (n = 30)

Day 1
(mean SE)

Day 7
(mean SE)

Day 14
(mean SE)

Day 21
(mean SE)

Day 28
(mean SE)

441.97 59.64
467.10 78.18
456.95 70.29

459.62 68.27
518.07 70.29

773.50 77.56**
948.09 148.05**

478.90 44.74
532.03 77.34

380.02 40. 48
318.07 49.54

p < 0.01, very significant.

**

Various studies in the past have linked tissue

damage caused by oxygen free radicals to
ischaemia-reperfusion mechanism (14, 17, 20).
Motivated by these, occasional attempts have been
made to study the role of oxidants when a fracture
is sustained. Gokturk et al (12) evaluated oxidant
status during bone healing in rats, by using MDA
levels in bone specimens as an indicator of oxidative stress. Statistically significant increases in
MDA levels were noted by them on days 7 and 14
after experimentally fracturing the right tibia of the
study rats. They concluded that oxidative stress
occurs during the 2nd and 3rd weeks after a fracture. The left tibia was used as the control leg in the
sense that it was pinned intramedullarly but not
fractured, considering this equivalent to a surgical
procedure. It was concluded that only pinning the
tibia did not increase the MDA levels. Since the
MDA levels were evaluated in the bone specimens,
they also postulated that this oxidative stress could
be a potential cause of impaired fracture healing.
In another recent study in a human population by
Basu et al (2), urinary levels of 8-iso-PGF2 a were
taken as a bio-marker of oxidative stress. Their aim
was to investigate whether oxidative stress had any
effect on the bone mass in humans. They used 15keto-dihydro-PGF2 a, an indicator of inflammatory response, as a control. Their findings established
a biochemical link between increased oxidative
stress and decreased bone density.
No previous study, to our knowledge, has been
carried out in humans to link fractures to oxidative
stress. This study was thus done as a preliminary
attempt to find any correlation and open avenues
for future research. We included only those fracture
patients who did not have any external soft tissue
injury, and excluded cases with associated injuries

like head injury (14, 27), blunt chest trauma (11, 20)
and blunt abdomen trauma (10), as these have been
implicated in oxidative stress and could have given
altered values.
To be able to include an appropriate sized cohort
of specimens in a short span of time, we expanded
the ages of our patient cohort to include cases ranging from 18 to 60 years of age. Additionally, to
avoid any exogenous source of antioxidants, we
ensured that no vitamin preparations were prescribed to these patients, as this could have potentially altered the values of the total endogenous
antioxidant status.
The ideal evaluation of oxidative stress in fracture healing would be from specimens collected
from the fracture site. In human conditions, it is
impractical and unjustified to take serial specimens
from the fracture site ; we therefore chose to use
plasma, which contains the by-products of any
reaction initiated due to oxygen free radical
induced tissue injury ; to our knowledge, this has
never been evaluated previously.
The increased values of RNI, cit and MDA by
7th day, peaking by the 14th day, and the decline
over the next two weeks showed a significant level
of oxidative stress during the 2nd and 3rd weeks
post fracture. The increase was significantly more
in group B than A, implying thereby that there is
more oxidative stress associated with multiple fractures, which seems to be logical.
The reason for the absence of recordable oxidative stress during the first week could be explained
by the fact that at the time of a fracture there is
interruption of blood flow to the severed bone and
plasma levels may not change initially. This regional ischaemia however leaves viable cells in the
peripheral regions of this ischaemic zone, which
Acta Orthopdica Belgica, Vol. 69 - 6 - 2003

550

G. PRASAD, M. S. DHILLON, M. KHULLAR, O. N. NAGI

could be salvaged if managed adequately. If not,

these cells may undergo irreversible injury. This
period of ischaemia lasts for about a week. Further
during the healing process, callus begins to form
and new cells are laid down, which includes new
capillary vessels. This leads to increased vascularity in the region. There is also an increase in the
inflammatory cells which increases the production
of oxygen free radicals. These radicals may then
lead to oxidative injury as seen in other tissues (8, 9,
14, 15, 17, 20) with reperfusion. This mechanism of
oxidative injury is based on the conclusions of previous studies (9, 14, 15, 20) that have been carried out
on other tissues which had suffered oxidative
injury. The exact mechanism however needs to be
confirmed by taking bone tissue samples directly
from the fracture site. These then can be subjected
to a detailed histopathological and biochemical
analysis. In humans, however, it was not possible to
take serial biopsies as it was ethically unjustifiable.
The FRAP values, which represent the total
endogenous antioxidant status, also showed a similar pattern : i.e. a rise during the 7th and 14th days
and gradual decline by the end of the 4th week.
Although we have no clear explanation for this, we
presume that it may be an attempt of the body to
combat the rise in oxidant levels.
Since surgery requires direct handling of the
fracture site and may disturb the healing process,
we attempted to correlate the date of surgery with
the different parameters evaluated. Most of the
cases were operated during the 1st and 2nd weeks
post injury. However, we could not correlate the
surgical insult to the oxidative stress as the time
between injury and surgery varied, and often more
than one surgical procedure was done on different
days (Group B, multiple fractures).
Increased age (2, 6, 26) has been related to
osteopenia and increased oxidative stress. We could
not completely evaluate variation in oxidative
stress levels with age in the present study due to the
study sample being too small ; however our data
revealed no statistical correlation of age at injury
with the level of recordable oxidative stress.
Considering the whole study, a lot more needs to
be studied. More stringent conditions for patient
selection need to be applied. Ideally, similar fracActa Orthopdica Belgica, Vol. 69 - 6 - 2003

ture patterns need to be taken up for study, and perhaps operated by a similar mode of fixation by the
same surgeon. Factors like blood pressure, heart
status, diabetes, smoking, alcohol, body mass
index, occupation, etc., which have an influence on
the oxidative stress need to be taken into consideration. The timing of surgery needs to be fixed. If
possible, representative tissue (from bone) needs to
be analyzed for the oxidative stress. The effect of
antioxidants needs to be seen. It is also recommended that fracture cases should be studied for a
longer period of time, perhaps till union is
achieved, to see for further variations.
This preliminary study has attempted to create a
platform for further research. We have shown our
observations and accepted our shortcomings. We
hope that the findings of this study can be capitalised to unmask any hidden fact, which may help
in understanding fracture healing better.
REFERENCES
1. Adam AK. Antioxidant vitamins and the prevention of
coronary artery disease. Am Fam Physician 1999 ; 60 :
895-904.
2. Basu S, Michaelsson K, Olofsson H, Johansson S,
Methus H. Association between oxidative stress and bone
mineral density. Biochem Biophys Research Communications 2001 ; 288 : 275-279.
3. Beal MF. Aging, energy and oxidative stress in neurodegenerative diseases. Ann Neurol 1995 ; 38 : 357-366.
4. Benzie IFF, Strain JJ. FRAP Assay. Ann Biochem 1996 ;
239 : 70.
5. Boyde TRC, Rahmatullah M. Optimisation of conditions
for the calorimetric determination of citrulline using
diacetyl monoxime. Ann Biochem 1980 ; 107 : 424-431.
6. Bremner I, Aggett PJ, James P. Micronutrients in health
and disease. British Medical Bulletin Series 1999 ; 55 : 3.
7. Collins PF. Estimation of FRAP value. Ann Biochem
1959 ; 31 : 1862.
8. Cornell CN, Lane JM. Newest factors in fracture healing.
Clin Orthop 1992 ; 277 : 297-311.
9. Dhaliwal H, Kirshenbaum LA Randhawa AL,
Singhal PK. Correlation between antioxidant changes
during hypoxia and recovery on reoxygenation. Am J
Physiol 1991 ; 261 : 632-638.
10. Farber JL, Kyle ME, Coleman JB. Biology of disease,
mechanism of cell injury by activated oxygen species. Lab
Invest 1990 ; 62 : 670-679.
11. Girotti MJ, Khan N, Mc Lellan BA. Early management
of systemic lipid peroxidation products in the plasma of
major blunt trauma patients. J Trauma 1991 ; 31 : 32-35.

EVALUATION OF OXIDATIVE STRESS

12. Gokturk E, Turgut A, Baycu C, Gunal J, Seber S,

Gulbas Z. Oxygen free radicals impair fracture healing in
rats. Acta Orthop Scand 1995 ; 66 : 473-475.
13. Green TK, Burk RF, Sies H. Estimation of nitrogen intermediates. Ann Biochem 1982 ; 126 : 131-138.
14. Ikeda Y, Anderson JH, Long DM. Oxygen free radicals
in the genesis of traumatic and peritumoral brain edema.
Neurosurgery 1989 ; 24 : 679-684.
15. Kirshenbaum LA, Singhal PK. Increase in endogenous
antioxidant enzymes protects heart against reperfusion
injury. Am J Physiol 1993 ; 265 : 484-493.
16. Morrissey PM, OBrien NM. Dietary antioxidants in
health and disease. Int Dairy J 1998 ; 8 : 463-472.
17. Oda T, Nakai I, Mitou M, Yamagisi H, Oka T,
Yoshikawa T. Role of oxygen radicals and synergistic
effect of superoxide dismutase and catalase on ischemiareperfusion injury of the rat pancreas. Transplant Proc
1992 ; 24 : 797-798.
18. Parks DA, Bulkley GB, Granger DN, Hamilton SR, Mc
Cord JM. Ischaemia injury in the cat small intestine. Role
of superoxide radicals. Gastroenterology 1982 ; 82 : 9-15.
19. Raloff J. How antioxidants may fight cancer. Science
News 1959 ; 149 : 12.
20. Rangan U, Bulkley GB. Prospects for treatment of free
radical mediated tissue injury. Br Med Bull 1993 ; 49 :
700-718.

551

21. Sano M, Ernesto C, Thomas RG et al. A controlled trial

of selegiline, alpha-tocopherol, or both as treatment for
Alzheimers Disease. The Alzheimers Cooperative Study.
N Eng J Med 1997 ; 336 : 1216-1222.
22. Stocks K, Dormandy S. Lipid peroxidation estimation.
Indian J Biochem and Biophys 1971 ; 36 : 250.
23. Sugino K, Dohi K, Yamada K, Kawasaki T. The role of
lipid peroxidation in endotoxin-induced hepatic damage
and the protective effect of antioxidants. Surgery 1987 ;
101 : 746-752.
24. Symons MC. Radicals generated by bone cutting and fracture. Free Radic Biol Med 1996 ; 20 ; 831-835.
25. Titheradge MA. The enzymatic measurement of nitrate
and nitrite. In : Methods in Molecular Biology, vol. 100.
Nitric Oxide Protocols. Titheradge MA, ed. New Jersey :
Humana Press Inc, 1998 ; pp 83-90.
26. Varanasi SS, Francis RM, Berger C, Papika SS,
Datta HK. Mitochondrial DNA deletion associated oxidative stress and severe male osteoporosis. Osteoporosis Int
1999 ; 10 : 143-9.
27. Wildburger R, Borovic S, Zarkovic N, Tatzber F. Post
traumatic dynamic changes in the antibody titer against
oxidized low density lipoproteins. Wien Klin Wochenschr
2000 ; 112 : 798-803.

Acta Orthopdica Belgica, Vol. 69 - 6 - 2003